MaziyarPanahi/calme-2.2-qwen2-7b

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:Jun 27, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

Calme-2.2-Qwen2-7B is a 7 billion parameter causal language model developed by MaziyarPanahi, fine-tuned from the Qwen/Qwen2-7B base model. This iteration aims to enhance the base model's performance across various benchmarks, making it a general-purpose model suitable for a wide range of text generation tasks. It utilizes the ChatML prompt template for instruction following and is available in quantized GGUF formats for efficient deployment.

Loading preview...

Overview

MaziyarPanahi/calme-2.2-qwen2-7b is a fine-tuned version of the Qwen/Qwen2-7B model, developed by MaziyarPanahi. The primary goal of this fine-tuning effort is to achieve improved performance across a broad spectrum of benchmarks, making it a more capable general-purpose language model.

Key Capabilities

  • Enhanced General Performance: Aims to surpass the base Qwen2-7B model in overall benchmark scores.
  • Instruction Following: Utilizes the ChatML prompt template for clear and effective instruction-tuned interactions.
  • Quantized Versions Available: Offers GGUF quantized models for optimized inference and reduced memory footprint, accessible via the MaziyarPanahi/calme-2.2-qwen2-7b-GGUF repository.

Open LLM Leaderboard Performance

This model has been evaluated on the Open LLM Leaderboard, demonstrating its capabilities across several metrics. Key scores include:

  • Avg.: 23.23
  • IFEval (0-Shot): 35.97
  • BBH (3-Shot): 33.11
  • MMLU-PRO (5-shot): 32.21

Detailed results are available on the Open LLM Leaderboard.

Good for

  • General text generation and conversational AI applications.
  • Developers seeking a fine-tuned 7B parameter model with improved benchmark performance.
  • Use cases requiring efficient deployment, leveraging the provided GGUF quantized versions.